ui/insights_ui_generator.py

import logging
from typing import Dict, Any, List, Optional
import pandas as pd

# Configure logger for this module. Assumes logging is configured in app.py or main entry point.
logger = logging.getLogger(__name__)


def format_report_for_display(report_data: Optional[pd.Series]) -> str:
    """
    Generates a complete Markdown string for a single report, including a dynamic title
    based on the report's type and creation date.

    Args:
        report_data: A pandas Series representing a single row from the agentic analysis DataFrame.
                     It should contain 'comprehensive_analysis_text', 'report_type', and 'Created Date'.

    Returns:
        A Markdown formatted string for the report.
    """
    if report_data is None or report_data.empty:
        return "## Report Details\n\n*Please select a report from the library to view its contents.*"

    # Assuming the main report text is in a column named 'comprehensive_analysis_text'
    # You may need to adjust this column name based on your Bubble.io data structure.
    report_text = report_data.get('report_text', '*Report content not found.*')
    report_type = report_data.get('report_type')
    created_date_str = report_data.get('Created Date') # Bubble's default field name

    title = "Comprehensive Analysis Report"  # Default title

    try:
        if report_type == 'Quarte':  # As per your request for "Quarter" type
            title = "Quarterly Insights Report"
        elif report_type == 'Week' and pd.notna(created_date_str):
            # Bubble dates are typically in ISO format, e.g., '2024-06-11T14:30:00.000Z'
            created_date = pd.to_datetime(created_date_str)
            day_name = created_date.strftime('%A')  # e.g., 'Tuesday'
            title = f"{day_name}'s Weekly Update Report"
    except Exception as e:
        logger.error(f"Error generating dynamic report title: {e}")
        # In case of an error, the default title will be used.

    return f"## {title}\n\n{report_text.strip()}"


def format_report_to_markdown(report_string: Optional[str]) -> str:
    """
    Formats the comprehensive analysis report string into a displayable Markdown format.
    This can be enhanced to add more structure if the report has implicit sections.

    Args:
        report_string: The raw text report from the orchestrator.

    Returns:
        A Markdown formatted string.
    """
    if not report_string or not report_string.strip():
        return "## Comprehensive Analysis Report\n\n*No analysis report was generated, or an error occurred during its generation.*"
    
    # Simple formatting for now. Could be enhanced (e.g., looking for patterns like "Section X:" to make them H3)
    # Ensure paragraphs are separated. Replace multiple newlines with double newlines for Markdown paragraphs.
    # report_string_cleaned = re.sub(r'\n\s*\n', '\n\n', report_string.strip())
    
    formatted_report = f"## Comprehensive Analysis Report\n\n{report_string.strip()}"
    # You might add more sophisticated parsing here if your LLM output for the report
    # has a consistent structure that can be converted to richer Markdown.
    return formatted_report

def extract_key_results_for_selection(
    actionable_okrs_and_tasks_dict: Optional[Dict[str, Any]]
) -> List[Dict[str, Any]]:
    """
    Extracts Key Results from the OKR structure for UI selection in Gradio.
    Each Key Result is given a unique ID for state management in the Gradio app.

    Args:
        actionable_okrs_and_tasks_dict: The dictionary representation of TaskExtractionOutput,
                                      typically `orchestration_results["actionable_okrs_and_tasks"]`.
                                      Expected structure: {'okrs': List[OKR_dict], ...}

    Returns:
        A list of dictionaries, where each dictionary represents a Key Result:
        {'okr_index': int, 'kr_index': int, 'okr_objective': str, 
         'kr_description': str, 'unique_kr_id': str}
    """
    key_results_for_ui: List[Dict[str, Any]] = []
    
    if not actionable_okrs_and_tasks_dict or not isinstance(actionable_okrs_and_tasks_dict.get('okrs'), list):
        logger.warning("No 'okrs' list found or it's not a list in the provided task extraction output.")
        return key_results_for_ui

    okrs_list = actionable_okrs_and_tasks_dict['okrs']

    for okr_idx, okr_data in enumerate(okrs_list):
        if not isinstance(okr_data, dict):
            logger.warning(f"OKR item at index {okr_idx} is not a dictionary, skipping.")
            continue
            
        okr_objective = okr_data.get('objective_description', f"Objective {okr_idx + 1} (Unnamed)")
        key_results_list = okr_data.get('key_results', [])

        if not isinstance(key_results_list, list):
            logger.warning(f"Expected 'key_results' in OKR '{okr_objective}' (index {okr_idx}) to be a list, got {type(key_results_list)}.")
            continue

        for kr_idx, kr_data in enumerate(key_results_list):
            if not isinstance(kr_data, dict):
                logger.warning(f"Key Result item for OKR '{okr_objective}' at KR index {kr_idx} is not a dictionary, skipping.")
                continue

            kr_description = kr_data.get('key_result_description') or kr_data.get('description') or f"Key Result {kr_idx+1} (No description)"
            key_results_for_ui.append({
                'okr_index': okr_idx,  # Index of the parent OKR in the original list
                'kr_index': kr_idx,    # Index of this KR within its parent OKR
                'okr_objective': okr_objective,
                'kr_description': kr_description,
                'unique_kr_id': f"okr{okr_idx}_kr{kr_idx}" # Unique ID for Gradio component linking
            })
            
    if not key_results_for_ui:
        logger.info("No Key Results were extracted for selection from the OKR data.")
        
    return key_results_for_ui

def format_single_okr_for_display(
    okr_data: Dict[str, Any], 
    accepted_kr_indices: Optional[List[int]] = None,
    okr_main_index: Optional[int] = None # For titling if needed
) -> str:
    """
    Formats a single complete OKR object (with its Key Results and Tasks) into a 
    detailed Markdown string for display. Optionally filters to show only accepted Key Results.

    Args:
        okr_data: A dictionary representing a single OKR from the TaskExtractionOutput.
        accepted_kr_indices: Optional list of indices of Key Results within this OKR 
                             that were accepted by the user. If None, all KRs are displayed.
        okr_main_index: Optional index of this OKR in the main list, for titling.


    Returns:
        A Markdown formatted string representing the OKR.
    """
    if not okr_data or not isinstance(okr_data, dict):
        return "*Invalid OKR data provided for display.*\n"

    md_parts = []
    
    objective_title_num = f" {okr_main_index + 1}" if okr_main_index is not None else ""
    objective = okr_data.get('objective_description') or okr_data.get('description') or f"Unnamed Objective{objective_title_num}"
    logger.info(f"OKR data desccr {objective}")
    objective_timeline = okr_data.get('objective_timeline', '')
    objective_owner = okr_data.get('objective_owner', 'N/A')

    md_parts.append(f"### Objective{objective_title_num}: {objective}")
    if objective_timeline:
        md_parts.append(f"**Overall Timeline:** {objective_timeline}")
    if objective_owner and objective_owner != 'N/A':
        md_parts.append(f"**Overall Owner:** {objective_owner}")
    md_parts.append("\n---")

    key_results_list = okr_data.get('key_results', [])
    displayed_kr_count = 0

    if not isinstance(key_results_list, list) or not key_results_list:
        md_parts.append("\n*No Key Results defined for this objective.*")
    else:
        for kr_idx, kr_data in enumerate(key_results_list):
            if accepted_kr_indices is not None and kr_idx not in accepted_kr_indices:
                continue # Skip this KR if a filter is applied and it's not in the accepted list
            
            displayed_kr_count +=1

            if not isinstance(kr_data, dict):
                md_parts.append(f"\n**Key Result {kr_idx+1}:** *Invalid data format for this Key Result.*")
                continue

            kr_desc = kr_data.get('key_result_description') or kr_data.get('description') or f"Key Result {kr_idx+1} (No description)"
            logger.info(f"KR data desccr {kr_desc}")
            target_metric = kr_data.get('target_metric')
            target_value = kr_data.get('target_value')
            kr_data_subj = kr_data.get('data_subject')
            kr_type = kr_data.get('key_result_type')

            md_parts.append(f"\n#### Key Result {displayed_kr_count} (Original Index: {kr_idx+1}): {kr_desc}")
            if target_metric and target_value:
                md_parts.append(f"  - **Target:** Measure `{target_metric}` to achieve/reach `{target_value}`")
            if kr_type and kr_data_subj:
                md_parts.append(f" **Key result type**: {kr_type}, for **data subject** {kr_data_subj}")

            
            tasks_list = kr_data.get('tasks', [])
            if tasks_list and isinstance(tasks_list, list):
                md_parts.append("  **Associated Tasks:**")
                for task_idx, task_data in enumerate(tasks_list):
                    if not isinstance(task_data, dict):
                        md_parts.append(f"    - Task {task_idx+1}: *Invalid data format for this task.*")
                        continue

                    task_desc = task_data.get('task_description') or task_data.get('description') or f"Task {task_idx+1} (No description)"
                    logger.info(f"task data desccr {task_desc}")
                    task_cat = task_data.get('task_category') or task_data.get('category') or 'N/A'
                    task_effort = task_data.get('effort', 'N/A')
                    task_timeline = task_data.get('timeline', 'N/A')
                    task_priority = task_data.get('priority', 'N/A')
                    task_responsible = task_data.get('responsible_party', 'N/A')
                    task_type = task_data.get('task_type', 'N/A')
                    data_subject_val = task_data.get('data_subject')
                    data_subject_str = f", Data Subject: `{data_subject_val}`" if data_subject_val and task_type == 'tracking' else ""

                    md_parts.append(f"    - **{task_idx+1}. {task_desc}**")
                    md_parts.append(f"      - *Category:* {task_cat} | *Type:* {task_type}{data_subject_str}")
                    md_parts.append(f"      - *Priority:* **{task_priority}** | *Effort:* {task_effort} | *Timeline:* {task_timeline}")
                    md_parts.append(f"      - *Responsible:* {task_responsible}")
                    
                    obj_deliv = task_data.get('objective_deliverable')
                    if obj_deliv: md_parts.append(f"      - *Objective/Deliverable:* {obj_deliv}")
                    
                    success_crit = task_data.get('success_criteria_metrics')
                    if success_crit: md_parts.append(f"      - *Success Metrics:* {success_crit}")

                    why_prop = task_data.get('why_proposed')
                    if why_prop: md_parts.append(f"      - *Rationale:* {why_prop}")
                    
                    priority_just = task_data.get('priority_justification')
                    if priority_just: md_parts.append(f"      - *Priority Justification:* {priority_just}")

                    dependencies = task_data.get('dependencies_prerequisites')
                    if dependencies: md_parts.append(f"      - *Dependencies:* {dependencies}")
                    md_parts.append("") # Extra newline for spacing between tasks details
            else:
                md_parts.append("  *No tasks defined for this Key Result.*")
            md_parts.append("\n---\n") # Separator between Key Results
    
    if displayed_kr_count == 0 and accepted_kr_indices is not None:
        md_parts.append("\n*No Key Results matching the 'accepted' filter for this objective.*")

    return "\n".join(md_parts)